Annals of Epidemiology 24 (2014) 720e726
Contents lists available at ScienceDirect
Annals of Epidemiology journal homepage: www.annalsofepidemiology.org
Original article
Ethnicity and alcohol consumption among US adults with diabetes Patrice A.C. Vaeth DrPH a, b, *, Raul Caetano MD, PhD c, Eva M. Durazo MPH a, b a
Prevention Research Center, Oakland, CA Pacific Institute for Research and Evaluation, Calverton, MD c University of Texas School of Public Health, Dallas Regional Campus, Dallas b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 March 2014 Accepted 21 July 2014 Available online 30 July 2014
Purpose: The drinking practices of a nationally representative sample of white, black, Mexican American, and other Hispanic adult diabetics are described and compared. Methods: Combined years (2005/2006e2011/2012) of the National Health and Nutrition Examination Survey provided home interview data from 2220 adults with self-reported diabetes of white (n ¼ 875), black (n ¼ 720), Mexican American (n ¼ 402), and other Hispanic (n ¼ 223) ethnicity. Current drinking status, the number of drinks consumed per week, and binge drinking were compared across ethnicity. Results: The multivariate findings for both diabetic men and women showed no statistically significant ethnic differences in current drinking status, and among women, there were no statistically significant ethnic differences in binge drinking. Among male diabetics, Mexican Americans consumed more drinks per week than whites (b ¼ 0.35; 95% confidence interval, 0.13e0.58; P ¼ .002) and were at increased risk for binge drinking (odds ratio, 2.04; 95% confidence interval, 1.30e3.21; P ¼ .002). Conclusions: Binge drinking is prevalent among Mexican American male diabetics. This pattern of drinking may put them at risk for poor diabetes management and control. It is important that health care providers routinely assess their patients’ drinking practices and address the health risks associated with alcohol consumption. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Diabetes Alcohol consumption Ethnic differences
Introduction Diabetes is a devastating disease that disproportionally affects many minority groups relative to whites [1e3]. Existing research indicates that alcohol consumption may impact diabetes management and control [4e7]. Little is known, however, about the drinking practices of those with diabetes and particularly, how this varies in relation to ethnicity. This is important to assess because US general population studies have documented significant differences across ethnicity in both the patterns of alcohol consumption and its adverse effects [8e16]. For example, among men, whites and Hispanics were more likely than blacks to be current drinkers, but blacks were more likely than whites and Hispanics to be daily heavy drinkers (5 drinks/d). White women were also more likely to be current drinkers compared with Hispanic and black women. Black and white women were more likely than Hispanic women to be daily heavy drinkers (4 drinks/d) [10]. There is also heterogeneity in drinking among US Hispanics. Mexican Americans and Puerto Ricans were more likely to be current drinkers, had higher weekly consumption levels, and were more likely to be binge drinkers than * Corresponding author. Prevention Research Center, 180 Grand Avenue, Suite 1200, Oakland, CA 94612-3749. Tel.: þ1 (510) 883 5759; fax: þ1 (510) 833 6440594. E-mail address: [email protected] (P.A.C. Vaeth). 1047-2797/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.annepidem.2014.07.007
Cuban Americans or South or Central Americans [13]. It is unclear, however, if similar ethnic-specific drinking practices exist among those with diabetes. This article describes and compares the drinking practices (current drinking status, number of drinks consumed per week, and binge drinking) of a nationally representative sample of white, black, Mexican American, and other Hispanic adults with diabetes. Based on ethnic differences in drinking practices in the general population (described previously), we hypothesize that white diabetics will be more likely to be current drinkers and will consume a higher number of drinks per week than black, Mexican American, and other Hispanics. In addition, among men, we expect that Mexican American diabetics will be more likely than white, black, and other Hispanic diabetics to be binge drinkers. Among women, we expect white diabetics to be more likely than black, Mexican American, and other Hispanic diabetics to be binge drinkers. Methods Sampling and data collection Respondents are from four combined years (2005/2006e2011/ 2012) of the National Health and Nutrition Examination Survey (NHANES) [17]. This cross-sectional survey is conducted by the
P.A.C. Vaeth et al. / Annals of Epidemiology 24 (2014) 720e726
National Center for Health Statistics and provides nationally representative estimates of health and disease in the US civilian noninstitutionalized population. NHANES includes in-home interviews using a computer-assisted personal interview system and health examinations in mobile examination centers. NHANES uses a complex multistage probability sampling design with oversampling of subgroups in different years. Approximately 5000 individuals participate annually. The mean response rate across the United States for 2005/2006 to 2011/2012 was approximately 75%. These analyses include home interview data from 2220 self-reported diabetic adults (21 years) who were of white (n ¼ 875), black (n ¼ 720), Mexican American (n ¼ 402), or other Hispanic (n ¼ 223) ethnicity and for whom there were complete data on alcohol consumption. Diabetic status was determined with the following question: “Other than during pregnancy, have you ever been told by a doctor or other health professional that you have diabetes or sugar diabetes?”
Measures Drinking status Drinking status was categorized for bivariate analysis as (1) lifelong abstainers, (2) ex-drinkers, and (3) current drinkers (any alcohol in the past 12 months). For multivariate analysis, drinking status was treated as a dichotomy (abstainers vs. drinkers). Number of drinks consumed per week This continuous measure was assessed by combining the selfreported frequency and quantity of drinking wine, beer, and liquor in the previous 12 months (standard drink: 5 ounces of wine, 12 ounces of beer, or 1½ ounces of liquor). The raw values are presented in Table 2, and because the distribution of this variable was substantially skewed, it was log transformed for the regression models presented in Table 4. Binge drinking The National Institute of Alcohol Abuse and Alcoholism (NIAAA) defines binge drinking as consumption of more than five (men) or four or more (women) drinks in about a 2-hour period [18]. NHANES did not define binge drinking in this manner until survey year 2011/2012. For consistency across survey years included in these analyses, binge drinking is defined as per survey years 2006 to 2010: as consumption of five or more in a day in the previous year. Respondents were categorized as (1) lifelong abstainers, exdrinkers, and current drinkers who did not binge; and (2) current drinkers who binged one or more time. Binge drinking among current drinkers was also assessed: non-binge drinkers versus binge drinkers.
721
Table 1 Sample description: all diabetic respondents stratified by sex among those aged 21 years or older with self-reported diabetes (NHANES 2005e2012)*
Mean age (range: 21e85 y)
All diabetics n ¼ 2220
Male diabetics n ¼ 1113
Female diabetics n ¼ 1107
59.7 y (0.43)
59.3 y (0.58)
60.1 y (0.57)
% Ethnicity White Black Mexican American Other Hispanic Level of education
Contents lists available at ScienceDirect
Annals of Epidemiology journal homepage: www.annalsofepidemiology.org
Original article
Ethnicity and alcohol consumption among US adults with diabetes Patrice A.C. Vaeth DrPH a, b, *, Raul Caetano MD, PhD c, Eva M. Durazo MPH a, b a
Prevention Research Center, Oakland, CA Pacific Institute for Research and Evaluation, Calverton, MD c University of Texas School of Public Health, Dallas Regional Campus, Dallas b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 March 2014 Accepted 21 July 2014 Available online 30 July 2014
Purpose: The drinking practices of a nationally representative sample of white, black, Mexican American, and other Hispanic adult diabetics are described and compared. Methods: Combined years (2005/2006e2011/2012) of the National Health and Nutrition Examination Survey provided home interview data from 2220 adults with self-reported diabetes of white (n ¼ 875), black (n ¼ 720), Mexican American (n ¼ 402), and other Hispanic (n ¼ 223) ethnicity. Current drinking status, the number of drinks consumed per week, and binge drinking were compared across ethnicity. Results: The multivariate findings for both diabetic men and women showed no statistically significant ethnic differences in current drinking status, and among women, there were no statistically significant ethnic differences in binge drinking. Among male diabetics, Mexican Americans consumed more drinks per week than whites (b ¼ 0.35; 95% confidence interval, 0.13e0.58; P ¼ .002) and were at increased risk for binge drinking (odds ratio, 2.04; 95% confidence interval, 1.30e3.21; P ¼ .002). Conclusions: Binge drinking is prevalent among Mexican American male diabetics. This pattern of drinking may put them at risk for poor diabetes management and control. It is important that health care providers routinely assess their patients’ drinking practices and address the health risks associated with alcohol consumption. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Diabetes Alcohol consumption Ethnic differences
Introduction Diabetes is a devastating disease that disproportionally affects many minority groups relative to whites [1e3]. Existing research indicates that alcohol consumption may impact diabetes management and control [4e7]. Little is known, however, about the drinking practices of those with diabetes and particularly, how this varies in relation to ethnicity. This is important to assess because US general population studies have documented significant differences across ethnicity in both the patterns of alcohol consumption and its adverse effects [8e16]. For example, among men, whites and Hispanics were more likely than blacks to be current drinkers, but blacks were more likely than whites and Hispanics to be daily heavy drinkers (5 drinks/d). White women were also more likely to be current drinkers compared with Hispanic and black women. Black and white women were more likely than Hispanic women to be daily heavy drinkers (4 drinks/d) [10]. There is also heterogeneity in drinking among US Hispanics. Mexican Americans and Puerto Ricans were more likely to be current drinkers, had higher weekly consumption levels, and were more likely to be binge drinkers than * Corresponding author. Prevention Research Center, 180 Grand Avenue, Suite 1200, Oakland, CA 94612-3749. Tel.: þ1 (510) 883 5759; fax: þ1 (510) 833 6440594. E-mail address: [email protected] (P.A.C. Vaeth). 1047-2797/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.annepidem.2014.07.007
Cuban Americans or South or Central Americans [13]. It is unclear, however, if similar ethnic-specific drinking practices exist among those with diabetes. This article describes and compares the drinking practices (current drinking status, number of drinks consumed per week, and binge drinking) of a nationally representative sample of white, black, Mexican American, and other Hispanic adults with diabetes. Based on ethnic differences in drinking practices in the general population (described previously), we hypothesize that white diabetics will be more likely to be current drinkers and will consume a higher number of drinks per week than black, Mexican American, and other Hispanics. In addition, among men, we expect that Mexican American diabetics will be more likely than white, black, and other Hispanic diabetics to be binge drinkers. Among women, we expect white diabetics to be more likely than black, Mexican American, and other Hispanic diabetics to be binge drinkers. Methods Sampling and data collection Respondents are from four combined years (2005/2006e2011/ 2012) of the National Health and Nutrition Examination Survey (NHANES) [17]. This cross-sectional survey is conducted by the
P.A.C. Vaeth et al. / Annals of Epidemiology 24 (2014) 720e726
National Center for Health Statistics and provides nationally representative estimates of health and disease in the US civilian noninstitutionalized population. NHANES includes in-home interviews using a computer-assisted personal interview system and health examinations in mobile examination centers. NHANES uses a complex multistage probability sampling design with oversampling of subgroups in different years. Approximately 5000 individuals participate annually. The mean response rate across the United States for 2005/2006 to 2011/2012 was approximately 75%. These analyses include home interview data from 2220 self-reported diabetic adults (21 years) who were of white (n ¼ 875), black (n ¼ 720), Mexican American (n ¼ 402), or other Hispanic (n ¼ 223) ethnicity and for whom there were complete data on alcohol consumption. Diabetic status was determined with the following question: “Other than during pregnancy, have you ever been told by a doctor or other health professional that you have diabetes or sugar diabetes?”
Measures Drinking status Drinking status was categorized for bivariate analysis as (1) lifelong abstainers, (2) ex-drinkers, and (3) current drinkers (any alcohol in the past 12 months). For multivariate analysis, drinking status was treated as a dichotomy (abstainers vs. drinkers). Number of drinks consumed per week This continuous measure was assessed by combining the selfreported frequency and quantity of drinking wine, beer, and liquor in the previous 12 months (standard drink: 5 ounces of wine, 12 ounces of beer, or 1½ ounces of liquor). The raw values are presented in Table 2, and because the distribution of this variable was substantially skewed, it was log transformed for the regression models presented in Table 4. Binge drinking The National Institute of Alcohol Abuse and Alcoholism (NIAAA) defines binge drinking as consumption of more than five (men) or four or more (women) drinks in about a 2-hour period [18]. NHANES did not define binge drinking in this manner until survey year 2011/2012. For consistency across survey years included in these analyses, binge drinking is defined as per survey years 2006 to 2010: as consumption of five or more in a day in the previous year. Respondents were categorized as (1) lifelong abstainers, exdrinkers, and current drinkers who did not binge; and (2) current drinkers who binged one or more time. Binge drinking among current drinkers was also assessed: non-binge drinkers versus binge drinkers.
721
Table 1 Sample description: all diabetic respondents stratified by sex among those aged 21 years or older with self-reported diabetes (NHANES 2005e2012)*
Mean age (range: 21e85 y)
All diabetics n ¼ 2220
Male diabetics n ¼ 1113
Female diabetics n ¼ 1107
59.7 y (0.43)
59.3 y (0.58)
60.1 y (0.57)
% Ethnicity White Black Mexican American Other Hispanic Level of education
